Impact of Traffic Sign Diversity on Autonomous Vehicles

Abstract

Traffic sign classification is indispensable for road traffic systems, including automated ones. There is a fundamental difference in the visual appearance of traffic signs from one country to another. Each dataset has its design standards and regulations based on shape, color, and information content, making implementing classification and recognition techniques more difficult. This paper aims to assess the influence of traffic sign diversity on autonomous vehicles (AVs) by reviewing several previous studies, comparing, summarizing their results, and focusing on classifying and detecting traffic sign datasets based on color, shape, and deep learning spaces using various methods and applications. Furthermore, it covers the main challenges facing road designers and planners considering changes to road safety infrastructure. It will be argued that compiling and standardizing a comprehensive global database of traffic signs is very difficult because it is costly and complex in application. However, it is still one of the possible solutions for the coming decades. Recommendations for future developments are also presented in this study.

Strategic Data Navigation: Information Value-based Sample Selection

“…Deep Learning proved to be an outstanding tool for computer vision applications, starting with the MNIST dataset, that contains 60, 000 images of 28 × 28 grayscale images (Baldominos et al, 2019). Nowadays, NN models are able to handle complex real-life visual information, achieving high accuracy on various wellknown benchmark datasets (Voulodimos et al, 2018), such as CIFAR100 or ImageNet or also deal with special domain specific datasets (Aldoski and Koren, 2023).…”

Improving Autonomous Vehicle Perception through Evaluating LiDAR Capabilities and Handheld Retroreflectivity Assessments

“…[ 18 ]. To address these challenges, research has explored various methodologies for traffic sign recognition in AVs, including color-based categorization, shape-based detection, deep learning methods, and edge detection techniques [ 2 ]. Furthermore, investigations have been conducted to understand the impact of environmental factors, such as lighting, automotive paints, target angles, distances, and surface conditions, on the performance of LiDAR and cameras in AVs [ 19 ].…”

“…Traffic sign visibility is a complex and multifaceted phenomenon influenced by various factors, including environmental conditions, lighting, sign placement, and the perceptual capabilities of the system interpreting the signs [ 2 ]. Understanding traffic sign visibility is paramount for effectively designing and operating AV systems.…”

Video Compression Prototype for Autonomous Vehicles

“…It is necessary to transfer different types of information, and not all types of information can be transferred in a still image [2]; for example, the direction of drive can be figured out from a video but not from a still image. Speed is also information that can be figured out from video and not from a still image.…”